Electroconvulsive Therapy (ECT) is the most effective treatment in psychiatry, and among the most effective in medicine. Despite its apparent non-focal effects leading to a generalized seizure, its therapeutic benefits are specific to a few clinical syndromes, including major depressive disorder (MDD) and bipolar depression (BD). These two syndromes share core deficits in reward processing (i.e. anhedonia). ECT improves anhedonia across mood disorders and syndromes, implying selective effects on the functional dynamics and structural properties of reward networks. Reward-related functions represent key behavioral dimensions of pathological relevance across neuropsychiatric disorders, and have a central place as positive valence constructs in the RDoC matrix. There has been a growing recognition that ?anhedonia? does not represent a unitary dimension; among its subcategories, three constructs emerge with clear relevance to behavior and disease: consummation (liking), motivation (wanting) and reinforcement (learning). Quantitative behavioral measures exist for each of these three, with clinical validity as biomarkers and predictors of response. The anatomy of the reward network is well known, with a core in the ventral tegmental area (VTA) and the Nucleus Accumbens (NAc), and projections to cortical and subcortical nodes via the mesocorticolimbic pathway and its ramifications. The Human Connectome Project (HCP) has significantly advanced the technologies for imaging brain connections in humans, accelerating innovation in the emerging field of Connectomics. Preliminary data from our group describes the feasibility of obtaining multimodal MRI measures of reward circuit biology (morphometry, tractography, functional connectivity) in patients undergoing ECT, and extracting clinically meaningful information to identify treatment targets and develop biomarkers and predictors. At a time when therapeutic research is stalled due to the absence of clear targets and useful biomarkers, understanding the mechanisms of our most effective treatments is a priority for our field. In this study, we propose a novel translational strategy that takes advantage of the high efficacy and fast response of ECT, and uses it to probe target engagement at the circuit level. With a systems neuroscience framework, in line with NIMH strategic priorities and the RDoC Initiative, we will focus on reward circuitry and its clinical dimensions across two clinical syndromes that are commonly treated with ECT: MDD and BD. We will use HCP multimodal MRI protocols combined with validated behavioral measures of reward constructs to assess patients before, during and after ECT, in addition to a cohort of matched healthy controls that will be imaged twice. This study is innovative in its proposal to combine ECT with multimodal MRI as a framework to study anhedonia transdiagnostically, with the translational aims to (1) discover treatment targets, (2) develop biomarkers and (3) identify predictors of response.